Microporous materials with pore size in the dimension of less than 2 nm are promising for a wide range of applications, in gas sorption and storage, gas separation, molecular sieves, catalysis, sensoring, and energy storage. The fabrication of microporous materials to membranes is further attractive for molecular separations because membrane separation technology is more energy-efficient compared to conventional molecular separation technologies, such as cryogenic distillation and absorption.
Ordered frameworks, such as zeolite (NPL 1-2, paper on zeolites), and metal-organic frameworks (MOFs) (NPL 3-6, paper on MOFs), are assembled from building blocks into homogeneous crystals with precisely defined porous-framework architecture. However, these crystalline frameworks are generally brittle and suffer from difficulty in manufacturing to large scale separation membranes.
In contrast, industrial, solution processable, selective membranes are made of densely packed polymers where molecules transport follows a solution-diffusion mechanism, and presents a trade-off between permeability and selectivity (NPL 7, Freeman, Upper bound, 1999), known as an upper bound (NPL 8, Robeson's upper bound, 1991, 2008). Conventional polymers pack efficiently with low free volume in the matrix, where the molecules are dissolved in the polymer and diffuse slowly through the free volume. Therefore, the molecular transport through these polymers is very slow, giving considerably low gas permeability that limits the large scale application of membrane separation processes.
According to the theoretical prediction by Freeman, there are two strategies to enhance both the permeability and selectivity of next-generation polymeric membrane materials: (i) improving the solubility selectivity (SA/SB), such as crosslinked poly(ethylene oxide) based membranes (NPL, 9); (ii) increasing the stiffness of polymer chains while maintaining large interchain spacing, such as thermal rearranged (TR) polymers (NPL10-11) and polymers of intrinsic microporosity (PIMs) (NPL 12-17), which are microporous materials with interconnected free volumes or micropores. The pores and channels are interconnected with size at molecular dimensions (<1 nm), the membrane would become a molecular sieve, allowing smaller gas molecules passing through the channels while blocking larger ones. Polymers with rigid macromolecular structures have high free volume and high permeability, such as poly(1-trimethylsilyl-1-propyne) (PTMSP) [NPL 18, Nagai et al, Poly[1-(trimethylsilyl)-1-propyne] and related polymers: Synthesis, properties and functions, Prog. Polym. Sci. 2001, 26, 721]. However, these highly permeable polymers have very poor selectivity that limits the practical applications of the polymers in molecular separations.
In the past decade, the materials and chemistry field have witnessed an explosive growth of interest in synthesis, design and fabrication of new-generation microporous materials, such as Polymers of intrinsic microporosity (PIMs) [NPL 12-17], covalent organic frameworks (COFs) [NPL 19-20], and porous organic cages (POCs) [NPL 21-23]. These porous organic materials are highly promising for molecular separations, gas storage, catalysis, organic molecular sieves and membrane materials.
Polymers of Intrinsic Microporosity (PIMs) are one class of microporous organic materials [NPL 12-17]. The concept of polymers of intrinsic microporosity was first invented by Budd and McKeown in 2002. The international patent WO2003000774 A1 describes the organic microporous network materials comprising a rigid 3-dimensional network of planar porphyrinic macrocycles in which pyrrole residues of adjacent macrocycles are connected by rigid linkers which restrain these adjacent macrocycles such that their porphyrinic planes are in a non-co-planar orientation. Preferred materials in accordance with the invention are phthalocyanine networks. These organic microporous materials are known as network PIMs.
Another invention by Budd and McKeown, international patent WO2005012397A2 and U.S. Pat. No. 7,690,514 B2, describes microporous organic macromolecules comprised of first generally planar species connected by rigid linkers having a point of contortion such that two adjacent first planar species connected by the linker are held in non-coplanar orientation, subject to the proviso that the first species are other than porphyrinic macrocycles. Preferred points of contortion are spiro groups, bridged ring moieties and sterically congested single covalent bonds around which there is restricted rotation. A typical synthetic approach of a representative PIM-1 polymer is shown in the drawing FIG. 1. Such non-network linear polymer chains are soluble in common organic solvents, so they can processed by solution casting methods to form films, coating on other materials, or manufactured to any shapes.
The backbone of PIMs polymers inhibits the free rotation or large scale conformational change and polymer chains cannot pack efficiently in the amorphous solid state, forming irregularly shaped free volume elements at molecular dimensions, as visualized in FIG. 2. The fractional free volume in PIM is sufficiently high that free volume elements are effectively interconnected, behaving like micropores (dimensions <2 nm). As probed by gas sorption, the PIMs polymers contain relatively high surface areas (400-1000 m2/g). Such high free volume in PIMs form unique hour-glass shaped structure behave like interconnected micropores, which allow high sorption capacity and rapid diffusion of molecules, while the ‘bottleneck’ or channels or gateways occupied by densely packed chain behave as sieves that could selectively screen molecules with different sizes or kinetic diameters.
International patent PCT/US2005/038195 describes the synthesis of covalently linked organic frameworks and polyhedra, known as COFs. The COFs materials are synthesized from organic building blocks linked by covalent bonds forming crystalline architecture with tuneable porosity at molecular dimensions. The organic building blocks could be tuned to tailor the structure, functionality, and materials properties. Though the pore structure of COFs could be well-defined forming ordered crystalline structure, these ordered porous COFs materials are generally not solution processable, limiting their processing and further fabrication into more useful membranes.
A particular promising application of these microporous polymers is fabrication of membrane for molecular separations. International Patent WO 2005/113121 (PCT/GB2005/002028) describes the formation of thin film composite (TFC) membranes by coating PIMs polymer onto a porous support membrane. The membranes were demonstrated as promising for gas separation, pervaporation (phenol/water, ethanol/water), propene and propane separation, and nanofiltration.
Amorphous polymer chains interact by relatively weak noncovalent interactions such as van der Waals forces or entanglements, and easily slide over each other. At the microscopic level, the amorphous nature of PIMs polymer chains results in a broad size distribution of free volume elements (4 to 10 Å) with different topologies existing in all the PIMs polymers, which compromises their separation performance, i.e. poor molecular selectivity, physical aging and plasticization. In particular, for industrially and environmentally important gases, such as separations of CO2/CH4 and hydrocarbons in natural gas industry, all of which have high solubility in glassy polymers. All of the existing PIMs polymers have only shown modest selectivity for CO2/CH4 separations. Therefore, tailoring the distribution, size, and architecture of channels and free volume elements is critical to achieve substantial increase of diffusivity selectivity via molecular sieving function.
International patent WO2010124359A1 disclose a method of preparing carboxylated Polymers of Intrinsic Microporosity (PIMs) by controlling the degree of hydrolysis of with tunable gas transport properties. The essential concept is hydrolyzing all or a portion of the —CN groups to form —COOH groups. International patent WO2010048694 (A1) disclose chemical structure and synthesis strategies for new PIMs polymers and their application for separation membranes. Another patent WO2011057384A1 further provides a tetrazole-containing polymer of intrinsic microporosity. A polymer of intrinsic microporosity (PIM-1) was modified using a “click chemistry” [2+3] cycloaddition reaction with sodium azide and zinc chloride to yield new PIMs containing tetrazole units (TZPIMs), with details report in a paper by Du et al (NPL 16). Recently, other similar works on transformation of nitrile groups in PIMs polymer membranes have been reported. Mason et al [NPL 24] prepared a thionated PIM-1 membrane by reaction of PIM-1 with phosphorous pentasulfide in the presence of sodium sulfite. However, most of these modifications introduce hydrogen bonding, e.g. functional groups with the dioxane linkages, rather than covalent crosslinking. Therefore, the rigidity of the network structure is not stable and the selectivity is not sufficiently high. For example, the performance of solubility-favored TZPIMs remains to be improved for separation of condensable gas molecules (CO2 and CH4), which only shows modest selectivity (˜15) for CO2/CH4 gas pair, though the CO2 permeability is as high as 2000-3000 Barrer.
When polymers are used for molecular separations, such as gas separation, organic solvent nanofiltration and ethanol/water separations, the stability of polymer in chemicals and solvents becomes a critical issue. The desirable strategy is covalently crosslinking of the polymer chains to form network structure, so the polymer would become stable in practical applications. Covalent cross-linking of polymer is the process of chemically linking polymer chains by covalent bonds. After covalent crosslinking, the polymer molecules cannot slide over each other so easily and the resulting polymer network becomes tougher and less flexible. Such crosslinked network may become more resistant to plasticization by condensable gas molecules, therefore covalent crosslinking is a favourable approach to stabilize polymeric membranes.
United States patent U.S. Pat. No. 7,758,751 B1 claimed UV-cross-linked membranes from polymers of intrinsic microporosity (PIMs) and the use of such membranes for separations, which were prepared by exposing PIMs membranes to short-wavelength UV irradiation. U.S. Pat. No. 7,758,751 B1 patent claimed the crosslinking of polymer without any scientific proof. Another US patent US 20130247756 A1 claimed UV-rearranged PIM-1 membranes by similar process of UV irradiation and their use for hydrogen separation. In both patents, the UV irradiated membranes apparently show higher selectivity and lower permeability. However, the attribution of the performance to crosslinking (in the U.S. Pat. No. 7,758,751 B1 patent) or chain rearrangement (US 20130247756 A1) is clearly in error. Recent independent research by the inventors of this patent (NPL 25, Song et al, Nature Communications, Photo-oxidative enhancement of polymeric molecular sieve membranes, 2013) demonstrate that observed apparently enhanced selectivity in gas separation arises from the ultraviolet-induced photo-oxidation at the surface of the membranes, which produces a dense selective surface skin to a high-permeability and porous material. The photo-oxidation reactions of PIM-1 polymer induce oxidative chain scission at the surface of membranes rather than covalent crosslinking or rearrangement. Such photodegradation at surface is evidenced by loss in molecular weight and limited transmission of UV light within only several hundreds of nanometers at the surface, and solubility of the surface in polar solvents and organic vapour.
Thermal treatment is a commonly used method for processing of polymeric materials. However, thermal treatment at excessive high temperature could induce chemical reactions or degradation of the polymers depending on the thermal stability of functional groups and atmosphere. Degradation of polymers is defined as the deterioration of the properties, such as hardening, reduced ductility and brittleness, softening, cracking, colour changes, and reduction in other desirable physical properties. Thermal degradation can be divided into three types: depolymerisation, random chain scission, and substituent reactions. The former two types of degradation usually involve the scission of backbone and change of molecular weight (and distribution). On the one hand, significant degree of chemical reactions involved in thermal degradation at high temperature could lead to the deterioration of physical and chemical properties. On the other hand, thermal treatment in controlled atmospheres could also be useful for modifying the physical and chemical properties of polymer to achieve specific purpose, such as the enhanced molecular sieving function as reported in this invention.
Recently, covalently crosslinked PIMs have been prepared by mixing the PIM polymer with crosslinking agent that induce covalent crosslinking reactions upon heat treatment. For example, Du et al reported cross-linked PIMs membranes, prepared by a nitrene reaction from PIM-1 in the presence of two different diazide cross-linkers [NPL 26, N. Du et al., Azide-Based Cross-Linking of Polymers of Intrinsic Microporosity (PIMs) for Condensable Gas Separation. Macromol. Rapid Commun. 32, 631-636 (2011)1. The PIM-1 polymer was mixed with azides and covalently crosslinked by heat treatment at 175° C. for 7.5 h. These covalently cross-linked polymeric membranes showed improved selectivity in gas separation and higher resistance to CO2 plasticization. However, these crosslinking methods also result in significant loss in permeability, for example, the azide-crosslinked membranes show CO2 permeability of about 200-600 Barrer with the selectivity of CO2/N2 increase to 27.
Thermal crosslinking at intermediate temperature (<450° C.) are also used in crosslinking polymers. For example, Koros and co-workers [NPL 27, Kratochvil, A. M. & Koros, W. J., Decarboxylation-Induced Cross-Linking of a Polyimide for Enhanced CO2 Plasticization Resistance. Macromolecules 41, 7920-7927 (2008)] found the crosslinking of carboxylic acid-containing 6FDA-based polyimides annealed at temperature close to 375° C. The authors performed various characterization techniques and ruled out the possible mechanisms including charge transfer complexing, oligomer cross-linking, decomposition, and dianhydride formation. The decarboxylation reaction removes the pendant acid group creating a phenyl radical capable of attacking other portions of the polyimide forming covalent aliphatic and aryl crosslinking bonds. Similarly, Du et al [NPL 28, Du et al, Decarboxylation-Induced Cross-Linking of Polymers of Intrinsic Microporosity (Pims) for Membrane Gas Separation. Macromolecules 45, 5134-5139 (2012)] reported the decarboxylation-induced crosslinked PIM membranes by heat treatment of carboxylated PIMs membrane at relatively higher temperature (375° C.). The resulting crosslinked membranes show improved selectivity for several gas pairs, such as CO2/CH4 of 25 and CO2/N2 selectivity of 26 while the CO2 permeability maintained at reasonable high level at 1291 Barrer.
Recently, thermal treatment of PIM-1 membranes was reported by Li et al [NPL 29, F. Y. Li, Y. Xiao, T.-S. Chung, S. Kawi, High-Performance Thermally Self-Cross-Linked Polymer of Intrinsic Microporosity (PIM-1) Membranes for Energy Development. Macromolecules 45, 1427-1437 (2012)]. Li et al claimed the transformation of the nitrile groups of PIM-1 to triazine rings at an elevated temperature at 300° C. with a prolonged soaking time of 2 days under vacuum. Li et al reported both increase of gas permeability and selectivity with thermal treatment. The resulting crosslinked PIM-1 membrane thermally treated at 300° C. for 2 days had a CO2 permeability of 4000 barrer and a CO2/CH4 and CO2/N2 ideal selectivity of 54.8 and 41.7, respectively, which were far beyond the Robeson's upper bounds. While Li et al observed covalent crosslinking of PIM-1 polymers, however their proposed mechanism is questionable, and their uncontrolled operation conditions of thermal treatment, as well as results, are not reproducible and making it not possible to apply the method to control the properties of polymers.
Thermal treatment was also used to transform a class of polymers to microporous materials, known as thermally rearranged (TR) polymers. TR polymers are aromatic polymers prepared by thermal treatment of polyimides with ortho-functional groups (PIOFG) [NPL 10-11, Park et al, Science, 2007]. International patent WO2009113747 A1 and WO2012167114A2 disclose the details of the synthesis of thermally treated polymers and applications for gas separation membranes. International patent WO2012166153A1 further disclose the applications of thermally rearranged (TR) polymers as membranes for ethanol dehydration. In TR polymers, ortho-functional group can be hydroxyl (—OH), thiol (—SH), and amine (—NH2) groups. The thermal treatments are usually performed at intermediate temperatures (350-450° C.) in inert atmosphere, after which the polymers are transformed to aromatic, infusible, and insoluble materials. Advanced characterization analyses of TR polymers indicate that the free volume or micropore and their size distributions could be tuned by varying the monomer structures of the precursor polymers and by using different thermal treatment protocols. The thermal treatment transforms dense glassy polymer precursors with low free volume, to microporous materials with interconnected microcavities with a narrow cavity size distribution. Such change of microporous structure results in highly permeable and selective membranes with exceptional gas transport properties, especially in sieving light gas molecules from large molecules. For example, the CO2 permeability of a representative TR polymer is as high as 1600 Barrer while the CO2/CH4 selectivity is stable at about 50 with negligible plasticization effect at high pressure. In all of these TR process, the heat treatment were prepared in inert atmosphere, and the role of oxygen was not studied and poorly understood.
Heat treatment of polymer at high temperature (>500° C.) in inert atmosphere result in pyrolysis of polymer to carbon materials. International patent WO2011053403 A1 and articles by Koros and co-workers [NPL 30-31, papers by Koros and coworkers] discloses a process of controlled pyrolysis of polymer membranes by adjusting the concentration of oxygen in the pyrolysis atmosphere, to generate a microporous carbon molecular sieves (CMS) membranes for gas separation. The CMS membranes have tuneable gas separation performance in terms of selectivity and permeability that surpasses the upper bound of polymeric membranes. However, the CMS membranes have limited mechanical properties due to thermal treatment at high temperature (>500° C.).
Mixed matrix membranes (MMMs) or nanocomposite membranes are prepared by incorporation of molecular sieves or nanoparticles into polymer matrix. MMMs or nanocomposite membranes have been an active research area. A promising system is incorporation novel metal-organic frameworks crystals into polymer matrix [NPL 32-34]. Alternatively, non-porous inorganic nanoparticles could also be used a fillers for polymer nanocomposites [NPL 35]. International patent WO2007084169 A2 and US patent US 20070137477 A1 reported a method, composition and apparatus for forming a nanoparticle filled polymer having similar gas selectivity and higher gas permeability than the native polymers. The nanoparticles fillers are dispersed in polymeric materials to increase the permeability of the composite materials. International patent WO2007106677 A2 (application number PCT/US2007/063305) discloses the preparation of high flux mixed matrix membranes made by incorporating porous inorganic fillers (e.g. microporous and mesoporous molecular sieves, carbon molecular sieves, porous metal-organic frameworks) into PIMs for separations. However, the selectivity of these mixed matrix membranes are not significantly improved. For any type of fillers, either nanoparticles or porous molecular sieves, the properties of the polymer composites would largely depend on the properties of the polymer matrix itself. It is expected that covalently crosslinking of the polymer would also be effective to form the network structure filled by various types of fillers.
In summary, it is desirable to develop new processing methods to covalently crosslink novel microporous polymers while modifying and optimizing the structure of free volume elements or micropores to achieve much better molecular separation in terms of high permeability and selectivity that satisfying practical membrane separations, to improve the stability to chemicals and solvents that would allow their use as high-performance adsorbents or separation of chemicals and solvents, and many other applications.